Objective: To determine if histamine release occurs in clinical sepsis.
Design: Prospective, controlled, clinical study.
Setting: Interdisciplinary intensive care unit and trauma ward.
Patients: Sepsis was confirmed in 20 patients (test group) by the criteria of the Veterans Administration Systemic Sepsis Cooperative Study Group (1987) and was verified by positive blood culture. In addition, patients were scored by the Elebute and Stoner Sepsis Score (1983), as modified by Dionigi et al (1985). A concomitant control group consisted of 20 postoperative patients with non-life-threatening trauma to the extremities and without signs of local or systemic infection.
Interventions: Observational study. Blood samples were collected for determination of plasma histamine concentrations in both groups at the time of study entry and on five succeeding days.
Measurements and main results: The patients were well matched, and the groups were not significantly different for all criteria known to influence histamine release. Comparison of the median values of each group on days 1 through 5 demonstrated significantly higher plasma histamine values in the test group on days 1 through 4, but these values were no longer significantly higher on day 5. While none of the nonseptic control patients achieved a plasma histamine concentration of > 1 ng/mL (the concentration of which was considered to be the pathologic cutoff point representing histamine release), these values (i.e., > 1 ng/ mL) were found in nine of 20 test group patients. In the test group, nonsurvivors (n = 9) had significantly higher plasma histamine concentrations than survivors (n = 11) throughout the whole study and eight of nine nonsurvivors showed a plasma histamine concentration of > 1 ng/mL. Correlation of plasma histamine concentrations on day 1 to sepsis severity (initial Sepsis Score) showed that all but one patient with a combined low Sepsis Score (< 20 points) and histamine concentration of < 1 ng/mL survived, while all patients with a Sepsis Score of > 20 points and histamine release (plasma histamine concentration of > 1 ng/mL) died.
Conclusion: Increased histamine concentrations were shown to be causally associated (contributory determinant) with sepsis.